All incoming auditory, visual and somatosensory information is relayed through the thalamus to the cerebral cortex, which, during wakefulness, is responsible for transforming these sensory inputs into a spatio-temporal pattern of neuronal activity that gives rise to an internal representation of the external world. These representations are a crucial component of our conscious experience. Understanding how incoming information from the thalamus gives rise to cortical activity is therefore essential for understanding both normal and pathological states of consciousness, such as schizophrenia or autism, when internal representations appear to go awry. Although they are only a minority of all cortical neurons, GABA-releasing inhibitory interneurons play a crucial role in these transformations. By providing both feedforward and feedback inhibition, inhibitory interneurons restrict electrical activity in the majority excitatory neurons to a precisely tuned response in time and in space. Therefore, understanding how sensory representations are generated requires a detailed knowledge of the identity and properties of the GABAergic neurons and synapses which mediate feedforward and feedback inhibition, and of their responses to incoming sensory information, but such knowledge is still lacking. The two largest and best studied inhibitory subtypes are fast spiking interneurons and somatostatin-containing interneurons. Previous studies suggested that feedforward inhibition is mediated exclusively by fast-spiking cells. In contrast, we showed in the previous grant period that both subtypes of interneurons mediate feedforward inhibition, but do so under very different stimulation regimes: fast- spiking interneurons will fire in response to a transient stimulus, while somatostatin- containing interneurons will fire in response to a sustained input. During the current grant period, we will use electrophysiological and pharmacological methods to elucidate the biophysical basis for these striking differences in response properties, and we will use novel genetically modified strains of mice to test the roles of each interneuron subtype in shaping responses of cortical neurons to incoming sensory information from the thalamus.